What is LL-37?
LL-37 is the only cathelicidin-family antimicrobial peptide produced by the human body. It is a 37-amino-acid, alpha-helical fragment cleaved from a larger precursor protein called hCAP-18, the product of the CAMP gene. LL-37 research spans three decades and covers direct antimicrobial killing, immune signaling, and tissue repair, with more human trial data available than for most research peptides on this site.
Bengt Agerberth's group at the Karolinska Institute first described the precursor in 1995, naming the peptide FALL-39 for its first four residues (Agerberth et al. 1995, PNAS 92:195-199). Later work renamed the mature peptide LL-37, for its two starting leucines and 37-residue length. Unlike BPC-157 or TB-500, which are peptide sequences derived from other proteins, LL-37 is one the human immune system already makes and deploys against infection on its own.
How hCAP-18 is processed and what LL-37 does mechanistically
hCAP-18 is stored in the secondary granules of neutrophils and is also produced by keratinocytes, mast cells, and epithelial cells lining the airway and gut. It has no antimicrobial activity in its stored, unprocessed form. A 2001 study by Sorensen et al. in Blood (Sorensen et al. 2001, human neutrophil granule extracts) tested three serine proteases released from neutrophil azurophil granules and found that only proteinase 3 cleaved hCAP-18 into active LL-37 outside the cell. The other candidate proteases tested in the same experiments did not produce the mature peptide.
Once cleaved, LL-37 carries a net positive charge of about +6 and folds into an amphipathic helix, with hydrophobic residues on one face and charged residues on the other. That structure lets it insert into the negatively charged membranes of bacteria while mostly sparing the more neutral outer membranes of human cells. Proposed mechanisms of membrane disruption include a toroidal-pore model and a detergent-like carpet mechanism; which one dominates depends on peptide concentration and the target membrane's lipid makeup.
What LL-37 research shows about antimicrobial and immune effects
LL-37's antimicrobial spectrum is broad by peptide standards. In vitro assays across multiple laboratories report activity against gram-positive and gram-negative bacteria, including Escherichia coli, Pseudomonas aeruginosa, and methicillin-resistant Staphylococcus aureus, along with some fungi and enveloped viruses. Minimum inhibitory concentrations vary by organism and assay condition, typically in the low micromolar range, and physiological salt concentrations reduce potency against several strains.
Direct killing is only part of the picture. A 2003 study by Koczulla et al. in the Journal of Clinical Investigation (Koczulla et al. 2003, chick chorioallantoic membrane assay and rabbit hind-limb ischemia model) found that LL-37 induces new blood vessel growth through the formyl peptide receptor-like 1 receptor on endothelial cells. In the rabbit model, local LL-37 application increased capillary density and blood flow recovery after surgically induced limb ischemia, linking the peptide to arteriogenesis as well as host defense.
LL-37 also functions as an alarmin: a signal that recruits and activates immune cells beyond its direct antimicrobial effect. A 2007 study by Lande et al. in Nature (Lande et al. 2007, human plasmacytoid dendritic cell cultures) found that LL-37 binds self-DNA released from damaged cells and forms aggregates that plasmacytoid dendritic cells take up and deliver to intracellular TLR9 receptors. This triggers a strong interferon response, even though free self-DNA alone does not activate these cells.
The same mechanism is implicated in psoriasis, where excess LL-37 turns ordinarily inert self-DNA into a trigger for autoinflammation. A related 2007 study by Yamasaki et al. in Nature Medicine (Yamasaki et al. 2007, human skin biopsies and mouse model) found abnormally high cathelicidin peptide levels and elevated stratum corneum tryptic enzyme activity in rosacea-affected skin. Injecting the rosacea-associated peptide fragments into mouse skin reproduced the inflammatory phenotype, tying a single peptide to both a protective role and, when its processing goes wrong, a role in skin disease.
Despite this preclinical case, LL-37 has not become a licensed drug for infection or wounds. Two properties recur across the literature and work against it: the peptide is degraded quickly by host and bacterial proteases, and it turns cytotoxic to human cells at concentrations only modestly above those needed for antimicrobial activity. Both properties narrow the usable dose window, which is one reason the delivery studies below use local or topical application rather than systemic dosing.
Wound healing research and the human trial data
LL-37 concentrates at wound edges during normal healing, and blocking it experimentally slows repair. A 2003 organ-culture study by Heilborn et al. in the Journal of Investigative Dermatology found that neutralizing LL-37 with antibodies in human skin explants inhibited keratinocyte migration and reduced Ki67 proliferation-marker staining at the wound edge. That result indicates the peptide contributes directly to re-epithelialization rather than simply appearing as a bystander marker of injury.
A 2008 study by Carretero et al., also in the Journal of Investigative Dermatology (Carretero et al. 2008, HaCaT keratinocyte cultures and ob/ob diabetic mouse model), delivered LL-37 by adenoviral vector to excisional wounds in diabetic ob/ob mice, a standard model for impaired healing. Treated wounds showed significantly faster re-epithelialization and more granulation tissue than controls. The authors traced the effect to MAPK and PI3K-Akt pathway activation and increased expression of the Snail and Slug transcription factors that drive keratinocyte migration.
LL-37 is one of the few compounds on this site with completed human trial data. Gronberg et al. ran a randomized, placebo-controlled trial in patients with hard-to-heal venous leg ulcers (Gronberg et al. 2014, n=34, Wound Repair and Regeneration). After a three-week open-label placebo run-in period, participants received twice-weekly topical LL-37 at 0.5, 1.6, or 3.2 mg/mL, or placebo, for four weeks.
LL-37 was safe and well tolerated at every dose tested, and treated ulcers showed a greater reduction in wound area than placebo. The authors flagged the small sample size and short treatment window as limits on how far the result generalizes, and called for a larger confirmatory trial. None had been published as of mid-2026.
Practical considerations for research handling
LL-37 research peptide is supplied lyophilized and, like most peptides on this site, requires reconstitution before use. The step-by-step procedure for sterile technique, bacteriostatic water ratios, and injection volume calculation is covered in the peptide reconstitution guide. Volume math for a given concentration and dose is handled by the research dosing calculator.
LL-37's strong net positive charge makes it prone to binding plasticware and glass surfaces, which can silently reduce the concentration of a working solution. Published in vitro protocols typically use low-protein-binding tubes and, where cytotoxicity to cultured cells is a concern, dilute stock solutions immediately before the experiment rather than storing them at working concentration. This handling detail is specific to cationic peptides like LL-37 and does not apply the same way to every compound.
Temperature and humidity in Indonesia add another variable. Lyophilized peptide, kept sealed and desiccated at -20°C, stays stable far longer than any reconstituted solution left at Bali or Jakarta ambient temperatures of 28 to 33°C. Reconstituted LL-37 should be refrigerated at 2 to 8°C and used within the manufacturer's stated window; the lyophilized peptide storage guide covers the tropical-climate specifics in more detail. Zurich Biotech supplies LL-37 alongside its BPC-157 and TB-500 tissue-repair compounds, each with HPLC purity testing and a Certificate of Analysis; see the BPC-157 research overview for a comparison of tissue-repair mechanisms across the two peptide families.